To the Editor We read with interest the meta-analysis by Duan et al1 assessing differences in outcomes between programmed cell death 1 (PD-1) vs programmed cell death ligand 1 (PD-L1) inhibitors in the treatment of patients with cancer. We agree that this is a clinically significant question for which indirect comparison may play a role in the absence of head-to-head trials. However, there were limitations to this analysis that, despite its compliance with PRISMA guidelines, suggest the authors’ conclusions (ie, superiority in efficacy of PD-1 over PD-L1 inhibitors) are not well supported.